By Eric Berman1
Greater West Africa is undergoing significant turmoil. In the past five years, more than half of the region’s countries2 have experienced a coup attempt (a growing threat across the continent) or are confronting armed secessionist movements.3 While some countries have made progress in confronting the growing threat of armed violence, other states’ abilities to govern have deteriorated. The ready supply of illicit small arms and ammunition—and sometimes even heavy weapons systems—facilitates violent extremism and undermines the rule of law.
Current rhetoric and agendas divert both attention and scarce resources from making greater progress on combatting the illicit proliferation of arms to non-state armed groups across the region. The overthrow of Muammar Gaddafi led to large outflows of arms and combatants from Libya to neighbouring and nearby countries as well as those further afield. Weapons from that conflict zone—and other parts of the continent—are known to circulate on the black market for many years due to persistent demand, well entrenched clandestine networks and small arms’ durability. The trade is difficult to quantify given that it operates in the shadows. That said, claims that arms from Libya continue to fuel conflicts in the Lake Chad Basin region have more to do with diverting attention from actual sources of illicit arms flows than in highlighting meaningful current threats.
Moreover, officials of governments and regional organisations are quick to identify “porous borders” as undermining national security. And considerable ink is spilled to highlight the challenge and efforts to address it. Without dismissing these concerns or the need to evaluate various interventions, doing so must not dilute attention to, or resources for, the much greater problem addressed in this paper: the diversion of state arsenals to non-state armed groups within a country’s borders.
Understanding the map
The incidents displayed on the map above indicate the challenge governments—and governmental organisations—face when it comes to safeguarding state-owned weapons from enemy attack. More than a thousand assaults have occurred on uniformed personnel in the region between 2020 and 2024, with several hundred incidents likely resulting in the diversion of lethal materiel. Only 107 of these events4 are pictured here: 100 from the 18 countries within the region as well as 7 from Algeria and the Central African Republic (CAR) (as they too appear on the map and therefore deserve inclusion). These 107 attacks cover a broad cross section of perpetrators, targets, and situations, each of which is unpacked in turn below.
But first, some important caveats. The Safeguarding Security Sector Stockpiles (S4) Data Set consists of more than 2,500 records of attacks on security forces in more than 25 countries. Yet the database is still in its early stages of development and is very much a work in progress. It is not possible at this juncture to offer a truly representative sampling.5
- Temporal and geographic breadth emphasised: The incidents shown here were chosen primarily to illustrate temporal and geographic breadth. There is at least one event displayed for each of the 60 months covered between January 2020 and December 2024. Roughly half the incidents depicted occurred in 2020 and 2021 (23 and 32, respectively). (Events displayed from 2022 to 2024 number between 16 and 19 per year.) The priority was to identify a wide range of states, regions and provinces. Within those large administrative units a further filter was applied to cover many smaller subdivisions such as local government areas and departments.
- Full extent of losses only partially represented: Two additional methodological factors concerning levels of loss displayed merit mention. The severity of diversion encountered is likely understated as S4 errs on the side of caution when estimating the scale and scope of seizures.6 It is also noteworthy that states sometimes bomb or recover lethal materiel that armed groups have captured from them. The S4 Data Set records such events when known, although details of systems seized or destroyed are seldom shared. The categories attributed to incidents portrayed on the map do not reflect possible subsequent disablement or abandonment.
Finally, a word about the term “loss,” which carries very negative connotations with security forces. It is used here as a synonym for “seizure” and “diversion.” It is not meant to convey or confer culpability or lack of professionalism.
Several of the perpetrators are religiously or politically motivated and are affiliated with al Qaeda or the Islamic State, but most are not, and defy being pigeonholed into a single descriptive category. For example, entrepreneurial groups known as “bandits” and “kidnappers” across much of northwestern Nigeria number in the many dozens with members in the thousands. Ethnic groups, active in several states across the region, are spurred to violence due to deteriorating environmental conditions and threats to their livelihood. Moreover, broad categories such as “political” need to be unpacked: some groups want to overthrow the state, others wish to secede, and still others want their country’s democratically-elected government to be re-installed.
The targets of these groups include both state security actors and foreign forces serving either in peace operations or as part of bilateral security arrangements. State security actors involve primarily the army, gendarmerie and police. Non-state armed groups also have seized arms and ammunition from customs officers, forest and wildlife guards, prison wardens and even air force personnel, among others. The United Nations (UN) Multidimensional Integrated Stabilization Mission in Mali (MINUSMA) came under frequent attack and its troop- and police-contributing countries (T/PCCs) lost both personnel and lethal contingent-owned equipment (COE) to insurgents on numerous occasions. But the UN is only one of several actors that have authorised and undertaken peace operations in the region. Others include the African Union (AU), the Economic Community of West African States (ECOWAS), the European Union (EU), the Group of Five for the Sahel (G5 Sahel) and the Lake Chad Basin Commission (LCBC).7 The Table below provides a fuller listing and some additional information. Diversion of lethal materiel of T/PCCs has occurred in many of these actors’ missions. Countries that have concluded bilateral agreements and deployed armed uniformed personnel in one or more countries in greater Western Africa include France, Turkey, the Russian Federation, and the United States (US). (The United Arab Emirates has concluded contracts for military vehicles in several states across the region, and may have sent military trainers as well.) And, as mentioned above, foreign mercenaries and civilians armed by the state have also come under attack and lost weapons.
The types of seizures displayed range from assault rifles to main battle tanks. More than two-thirds of the incidents represent Category 1 events in the S4 Data Set: 10-49 firearms, 1,000-9,999 rounds of ammunition; or 1-4 gun trucks (i.e. a light, unarmoured, vehicle equipped with a machine gun) usually transporting several ammunition boxes (with capacities of 100-500 cartridges each). The Category 3 incidents (of which six are represented on the accompanying map) represent events in which 100 or more weapons, or 100,000 or more rounds of ammunition, or 10 or more gun trucks are seized. It is not uncommon for Category 3—and many Category 2—incidents to include the diversion of heavy weapons systems such as armoured vehicles as well as self-propelled and towed artillery.
Non-state armed groups engage their targets in a variety of settings. They have attacked fixed sites that include secure civilian and military airports, fortified checkpoints, outposts, forward operating bases, larger bases including sector and mission headquarters as well as training centres. Outside of these sites, state and international security forces are ambushed while performing escort duties, patrolling, conducting clearance operations, and as part of convoys or troop movements. Armed groups also seize weapons from uniformed personnel commuting to work and protecting people, such as VIPs at the office or at home, refugees at camps, or civilians shopping at markets. Forced abandonment due to fear or imminent attack also occurs on occasion (and almost certainly more often than recorded in open-source reporting). Corruption involving selling, bartering or gifting arms and ammunition to non-state armed groups is known to occur, but is difficult to gauge and almost impossible to geo-locate and date.
The S4 Data Set collates information concerning uniformed personnel targeted, types of attacks, small arms and heavy weapons systems diverted, as well as perpetrators. But such details are not distinguished on the map.
Some successes in safeguarding state-owned lethal materiel
Progress has been made. The three examples provided here are not meant to be taken as comprehensive. Nor are they necessarily the most important. Rather, they are meant to highlight efforts to address supply and demand issues, as well as management of existing stocks.
Nigerian security forces have lost considerably less lethal materiel to insurgents in recent years. Even without comprehensive data, this trend is clear. At least three factors explain this success. First, Abuja vastly reduced the number of targets. The army closed outposts and forward operating bases across Adamawa, Borno and Yobe and built or augmented a smaller number of bases, designated as Super Camps. (This practice protected troops and their equipment, but incurred other costs.) Second, security forces deployed enhanced aviation assets, including counter-insurgency aircraft and drones. This likely changed insurgents’ calculations concerning ground assaults, which could well be more difficult and costly with respect to loss of combatants and military hardware. And third, the two main jihadist factions have devoted time and energy fighting each other in the wake of the leadership crisis following Abubakar Shekau’s death in May 2021.
Côte d’Ivoire responded with alacrity and—it seems—effectively to attacks on its security forces near its border with Burkina Faso. The seeds of Yamoussoukro’s policy were planted after the 2016 terrorist incident against civilian targets in the coastal town of Grand-Bassem. The government redoubled its efforts after jihadists attacked an Ivorian security outpost in 2020. The country’s sizeable investments in economic development, social programmes, and enhanced deployments of uniformed personnel across its northern territories have apparently helped curtail a spate of attacks against both civilian and security targets in 2020 and 2021. Short-term success may not be a harbinger of what is to come. Moreover, replicating Côte d’Ivoire’s wide-ranging initiatives is not an option for many of the countries in the region whose economies are less robust. (Benin, for example, is experiencing greater challenges despite its determined efforts to counter the jihadist threat, and a windfall of Western support.) That said, it would be remiss not to flag and explore these efforts as well as the results to date.
Many investments—both foreign and domestic—made to enhance regional countries’ weapons and ammunition management (WAM) practices have paid off. It is always a challenge to prove a causal relationship to something that did not happen. Nevertheless, the destruction of many tons of surplus munitions8 and enhancements to numerous storage facilities arguably have helped reduce instances of materiel being diverted (or exploding). Provision of machines to mark state-owned and -authorised weapons—and training on how to use them—as well as ancillary record-keeping software provide important disincentives to sell, rent, or lose weapons that may be traced. Several EU member states—notably Germany—and the US have provided generous funding for these initiatives. Regional governments and ECOWAS have hired, trained, and equipped national bodies to implement this work as well as develop and promote good practice.
Some challenges to safeguarding state-owned lethal materiel
Challenges remain. The explanation for the selection of examples above applies to those below.
The ability of the UN to secure lethal COE during the closure of its peacekeeping mission in Mali was both impressive and problematic. It is not uncommon for organisations undertaking such deployments to face problems balancing mandate implementation with host-country expectations. MINUSMA was no exception. Even so, the final months of the 10-year operation were especially fraught. The UN was forced to close nine bases and withdraw 10,000-plus Blue Helmets—and their lethal equipment—under hazardous and inhospitable conditions, which Bamako exacerbated. Furthermore, logistical constraints dictated that the UN had to dismantle and disable substantial quantities of major equipment—including hundreds of armoured vehicles. While the UN stopped combatants from gaining access to this materiel, it also took serviceable armaments out of troop- and police-contributing countries’ arsenals. Depreciation and routine payments for wear and tear meant that some T/PCCs received far less remuneration from the UN than they had expected for their unexpected losses.9 This has arguably reduced the readiness of some T/PCCs from greater West Africa to respond to insurgencies back home and their willingness and ability to participate in future operations.
Countries’ oversight of arms and ammunition they provide to their citizens is cause for concern. Community security groups obtain weapons to defend themselves against insurgents and to supplement government efforts to uphold law and order. This can happen with or without government assistance. Such support can be sustained or episodic, and as a result of official national policy or informal practice at sub-national government levels. Arms provided may be rudimentary or sophisticated. Experience elsewhere—such as in Sierra Leone and in Sudan—underscores the perils that such measures have on the promotion of human rights and good governance. With Burkina Faso and Nigeria together recruiting and arming tens of thousands of civilians, the checks and balances in place to help ensure this materiel is not misused or misappropriated need considerably greater scrutiny than they currently face.
Regional arms control and WAM practices are unevenly implemented. Article 11 of the ECOWAS Small Arms Convention, for example, remains under-utilised when it should be extolled and robustly operationalised. On paper, Article 11 represents a global ‘best practice’. The Convention requires ECOWAS member states participating in peace operations to report what small arms and corresponding munitions they deploy with, what they resupply, what they recover, what they destroy, and what they take back home with them. A simple arithmetic equation in a spread sheet would document clearly what was used (either for training purposes or in implementation of the mission’s mandate), or could not otherwise be accounted for. This requirement is not limited to participation in ECOWAS peace operations. ECOWAS member states are active T/PCCs to numerous AU and UN—as well as other (e.g. LCBC)—missions, which adds to the ECOWAS Convention’s potential value. The Convention, concluded in 2006, has been legally binding since 2009. The ECOWAS Secretariat has adopted a revised reporting template for T/PCCs in its missions in the Gambia and Guinea-Bissau. That is an important start. More needs to be done.
Conclusion
As noted, the partial development of the S4 Data Set does not allow for broad trend analyses at present. It could be despite the persistent challenges states in the region face with regard to securing their stockpiles of lethal materiel, that governments, generally speaking, are doing a better job than was previously the case. Perhaps not. More data would answer this question, assess interventions’ successes, identify possible new priorities, and promote evidence-based good practice. The open-source information is there to be mined and rigorously assessed. Policymakers and program designers would both be well served to develop and utilise this tool.
Besides working to operationalise the ECOWAS Small Arms Convention, efforts should be undertaken to keep Burkina Faso, Mali and Niger engaged in existing processes and frameworks. There is no reason for these three countries to re-create the wheel as part of their newly created Alliance of Sahelian States (AES). While they are no longer obliged to abide by the ECOWAS Convention’s articles, there’s nothing to stop them from so doing. Mauritania, for example, has remained engaged in the West Africa Police Chiefs Committee (WAPCCO)—an ECOWAS institution—even though it formally withdrew from that regional bloc in 2000. In a similar vein, LCBC should be encouraged to take advantage of ECOWAS Convention best practice to the extent possible. Benin and Nigeria as T/PCCs to the Multinational Joint Task Force (MNJTF10), an LCBC operation, remain legally bound to report to the ECOWAS Secretariat in Abuja on their COE within the mission as concerns small arms and corresponding ammunition—as well as recovered materiel of the same nature. There is nothing prohibiting them from sharing this information with the LCBC Secretariat in N’Djamena. LCBC member states Cameroon and Chad could choose to do the same.
Arms control initiatives and frameworks should also be expanded to include weapon systems other than small arms. The diversion of state-owned gun trucks and various heavy weapon systems to non-state armed groups is a serious concern and not an uncommon occurrence. National and regional—as well as continental—bodies, agreements (whether legally or politically binding), and frameworks (such as the AU Silencing the Guns 2020 initiative, which the AU extended for ten years) should examine new threats as a lot has changed over the past 20-plus years.
Finally, governments in the region should be encouraged to share challenges as well as emerging good practice when it comes to arms control. This should cover oversight of both civilian and state weapons and arsenals. Policymakers and practitioners have amassed considerable experience. There is still a reluctance, however, to discuss what has worked—and what has not. The threats governments face are stark. Obfuscation and platitudes are unhelpful. Candor and improved practice are what is needed for greater progress to be made.
FURTHER READING
- Eric G. Berman is director of the Safeguarding Security Sector Stockpiles (S4) Initiative and a visiting scholar at Northwestern University’s Program of African Studies.
↩︎ - In this essay, “Greater West Africa” includes the 16 current and former members of the Economic Community of West African States (ECOWAS), as well as Cameroon and Chad. Former member states of that regional bloc include Burkina Faso, Mali, Mauritania, and Niger: Nouakchott formerly withdrew in 2000; Bamako, Niamey, and Ouagadougou formally withdrew in 2025.
↩︎ - Five countries in the region have experienced successful coup d’états since 2020: Burkina Faso (in 2022, twice), Chad (2021), Guinea (2021), Mali (2020 and 2021), and Niger (2023). Elected heads of state in the Gambia, Guinea-Bissau and Sierra Leone all withstood coup attempts in the 2022-2023 period. Further, in 2024, Benin claimed to have averted an attempted coup. Armed secessionist movements are active in Cameroon, Mali, Nigeria, and Senegal.
↩︎ - The events depicted may include attacks on civilians who are armed to protect their communities or whom the state has hired from abroad to support its effort to combat the threat armed groups pose, but only when they are co-deployed with state security actors, and never when they are attacked on their own.
↩︎ - Construction of the S4 Data Set to date has centred on greater West Africa, with a particular focus on the riparian states of LCBC (i.e. Cameroon, Chad, Niger, and Nigeria—and not CAR or Libya). In 2024, S4 and its partner the International Peace Information Service (IPIS) developed the database for Niger, expanding the collated information on the Diffa region to cover the country’s six other regions and capital district. In 2025, S4 and IPIS will concentrate on Cameroon’s Far North region, and examine incidents across the Southern African Development Community (SADC). Resources permitting, S4 and IPIS aspire to properly cover the African Union’s 55 member states to establish a functional baseline and allow for trend analyses, as well as promote good practice and accountability.
↩︎ - This does not mean there are no errors when categorising incidents. This can happen—and has happened. S4 revises its assumptions and summaries for events whenever better information becomes available. The clear trend over the past decade—since the Small Arms Survey published its 2015 study on attacks on peacekeepers in South Sudan and Sudan—is that better and fuller information tends to increase the number of recorded incidents and the levels of loss. S4 believes men and women in uniform merit greater support and protection in carrying out their legally sanctioned and crucially important duties. Better understanding how their arms are diverted to non-state armed groups, and reducing such loss are important goals that current approaches do not adequately address.
↩︎ - The Accra Initiative (AI) has authorised a peace operation, but has yet to field it as envisioned. Governments have shared information among the Initiative’s members, and a multinational headquarters team is operational in Tamale, Ghana. Deployments to date involve short joint military operations along members’ shared borders. The G5 Sahel joint force ceased operations in 2024, as did G5 Sahel.
↩︎ - In 2019 and 2020, just four countries—Burkina Faso, the Gambia, Mali, and Senegal—destroyed more than one million rounds of small arms ammunition and larger calibre munitions with international assistance.
↩︎ - Exceptionally, the destruction of ammunition belonging to T/PCCs at the end of the mission did not exacerbate this situation. Normally, T/PCCs would not be reimbursed for serviceable munitions that are destroyed, but an exception was made in this instance. (Author interview with knowledgeable source, 17 March 2025.)
↩︎ - On 29 March 2025, Niamey announced its decision to unilaterally withdraw from the MNJTF. ↩︎
This publication has been produced with the financial assistance of the Belgian Directorate-General for Development Cooperation and Humanitarian Aid (DGD). The contents of this document can under no circumstances be regarded as reflecting the position of the Belgian Development Cooperation.
